Biochemistry 1984, 23, 2983-2989 van den Broek, H. W. J., Nooden, L. D., Sevall, J. S.,& Bonner, J. (1973) Biochemistry 12, 229-236. Vytasek, R. (1982) Anal. Biochem. 120, 243-248. Welch, S . L., & Cole, R. D. (1979) J. Biol. Chem. 254, 662-665.
2983
Wu, C. (1980) Nature (London) 286, 854-860. Wu, C., Wong, Y.-C., & Elgin, S . C. R. (1979) Cell (Cambridge, Mass.) 16, 807-8 14. Yamamoto, K. R., & Alberts, B. (1975) Cell (Cambridge, MUSS.)4, 301-310.
Electronic Effects in the Acylation of a-Chymotrypsin by Substituted N-Benzoylimidazolest Robert L. Kogan and Thomas H. Fife*
ABSTRACT: Rate constants for the acylation of a-chymotrypsin by a series of acyl-substituted N-benzoylimidazoles have been determined by proflavin displacement from the active site. The second-order acylation rate constants k2/K, are large [e.g., that for N-(m-nitrobenzoy1)imidazoleis 1.7 X lo4 M-' s-l at pH 7.51, even though K, must be quite large (plots of k vs. k/[S], have infinite slopes). The values of k2/Kmare nearly independent of pH in the range 5.0-9.0 when the substituent group is electron donating. Electron-withdrawingsubstituents produce an increase in k2/K,,, with increasing pH until a maximum is reached near pH 7. This is also the case in acylation by the N- [p-(dimethylamino)benzoyl]-"-methylimidazolium ion (pKaPp= 6.5). While the reaction of the "-methylated derivative is via a positively charged species at all pH values, the unmethylated compounds react through both the neutral species and the conjugate acids, with the observed pH dependence depending on the relative values of the rate constants. The limiting value of k2/K, for the N-
[p-(dimethylamino)benzoyl]-N'-methylimidazolium ion is 2.1 times less in DzOthan in H 2 0 . Thus, His-57 must be participating in the acylation reaction as a general base. The limiting values of k2/Kmfor the corresponding "-methylated and unmethylated derivatives differ by a factor of only 150, which is similar to the difference in the second-order rate constants for nonenzymatic OH--catalyzed hydrolysis. A plot of log k2/K,,, vs. the Hammett substituent constant Q is linear with a slope p of 0.9 for rate constants obtained at pH 7.5 and nearly zero for rate constants determined at pH 8) for @-cyanobenzoyl)-a-chymotrypsin prepared s-l, whereas that from the acylimidazole was 3.89 X s-’. The obtained with thep-nitrophenyl ester was 3.92 X corresponding rate constants for deacylation of p-(dimethylamino)benzoyl-a-chymotrypsinwere 9.98 X 10” s-l and 9.14 x 10” s-*, respectively. It may be noted in Figure 2 that values of the rate constants (k2/K,) at pH 7.5 for I and the analogous ”-methylated derivative differ by only a factor of 150. This is also true in the nonenzymatic OH--catalyzed hydrolysis reactions. Plots of log koW for nonenzymatic hydrolysis at 30 OC and p = 0.1 M vs. pH are shown in Figure 4. Rate constants were determined in HCl solutions or in 0.01 M buffers (acetate, cacodylate, and Tris). There is buffer catalysis; however, it was shown that such low concentrations of buffer do not have an experimentally significant effect on the observed rate constants. The equation for kobsdin hydrolysis of N- [p-(dimethylamino)benzoyl]-N’-methylimidazolium ion is given in eq 3
+
k l a H 2 k,K,aH kobsd
=
OH2
+ koHK,Kw
+ K,aH
(3)
while that for hydrolysis of the unmethylated derivative I is given in eq 4, where k l and k2 are rate constants for attack
(4) of H 2 0 on the species with the p-(dimethylamino) group protonated and unprotonated, koH is the second-order rate constant for attack of OH- on the ”-methylated or protonated species of I having the p-(dimethylamino) group in the neutral, free-base form, koH’ is the second-order rate constant for attack of OH- on the neutral species of I, and K , is the dissociation constant for the p(dimethy1amino) group conjugate acid. The second-order rate constants koH’ and koH are 21.5 and 2150 M-’ s-’. A pH-independent reaction is encountered at pH 4-7 in hydrolysis of the N’-methyl derivative with k2 = 7.1 X s-l. At low pH, hydronium ion catalysis is observed with both compounds, which must be due to protonation of the p-(dimethylamino) group; kH = 1 M-’ s-’ ( k H
Discussion The acylation of a-chymotrypsin by N-benzoylimidazoles is experimentally a second-order reaction. Qualitatively similar second-order acylation reactions were also observed with aliphatic acylimidazole substrates (Kogan et al., 1982). In all cases, plots of k vs. k / [ S I ohad infinite slopes, which shows that if binding is occurring, then K , is much larger than the highest substrate concentration studied ( 10-2-10-3 M) so that ES is experimentally undetectable. Plots of k vs. [SI, were linear as shown in Figure 1 and give k 2 / K , as the slope. The constant k 2 / K , is the second-order rate constant for reaction of the free enzyme with substrate. The ratio k 2 / K , is not affected by any nonproductive binding of the substrate (Fersht, 1977) and is, therefore, the most reliable parameter to employ in structure-reactivity relationship studies (Brot & Bender, 1969). The acylation of the enzyme by the substituted Nbenzoylimidazoles with electron-donating substituents is nearly pH independent. Serine- 195 is undoubtedly being acylated in view of the identical rates of deacylation of acylchymotrypsins prepared from the N-acylimidazoles and the corresponding p-nitrophenyl esters. It is a reasonable assumption that K , is little affected by changes in pH since K , for nonionizing substrates has been shown to be essentially pH independent in the range 5-9 (Bender et al., 1964; Laidler & Barnard, 1956; Hammond & Gutfreund, 1955; Cunningham & Brown, 1956). The pH-rate constant profiles of Figure 2 must then primarily reflect the influence of pH on k2. The simplest kinetic scheme is one in which the N-acylimidazoles acylate a-chymotrypsin via both the protonated and neutral species or kinetic equivalents with rate constants that are not greatly different (eq 5) (Kogan et al., 1982). The pK, values E
EH+
+
S
e
S H+
ES
ESH*
k
products
products
of N-acylimidazoles and His-57 are approximately 4 and 7, respectively. Consequently, the scheme of eq 5 would explain the near pH independence of the experimental rate constants for I-IV in the pH range of 5-9 if the reactions of the protonated species are of major significance below pH 6 and the reactions of the neutral species are of greater importance at higher pH values.’ In the case of the p(dimethy1amino) derivative, k,‘/K,,, is 1 50-fold larger than k , / K , (assuming correspondence of k,‘/K, to the rate constant of the N’methylated compound). The reactions through the two species will therefore be approximately equal 2 pH units above the pK, of the N-acylimidazole, Le., at pH 6. Above this pH, the reaction of the neutral species will then predominate. The pH dependence of the rate constants will, of course, vary depending on the relative values of k, and k;, It will be noted in Figure I
The expression for k derived from the scheme of eq 5 is given in eq k / [ S l ~= (krK1K3 -+ ~ , ‘ K I ~ H ) / ( K I K+~KK&, , a H )
6, considering that K2 > aHand K ,
(6)
> [SI,. Therefore at pH 7, k,/K,,, would be of greatest importance if these terms are not greatly different [see the equations and discussion in Kogan et al. (1982)l.
ACYLATION OF a-CHYMOTRYPSIN
2 that when electron-withdrawingsubstituents are present, the rate constants increase with increasing pH to a maximum value near pH 7. This very likely reflects the increased difficulty of protonation and the increased reactivity of the neutral species. Electronic Effects in Acylation. The p value obtained from the plot of log (k2/K,) vs. u for acylation of a-chymotrypsin by the series of N-benzoylimidazoles at pH 7.5 (Figure 3) is only slightly positive (0.9). It should be noted that in view of the near pH independence of k2/Km,the p value is still 0.9 at higher pH, e.g., 8-8.5, where there can be no contribution from the reaction of the protonated species. The second-order rate constant for acylation is a composite constant, and substituent groups might therefore produce changes in either k2 or K,. However, the effects of substituents in the acyl group on the binding (K,) of substituted benzamides is quite small ( p = 0 to -0.3) (Marini & Caplow, 1971). Rate constants (kz/K,) for acylation of the enzyme by substituted benzoate esters having polar meta and para substituents were found to give reasonably good regression lines in Hammett plots, although there was marked positive deviation produced by nonpolar substituents, presumably due to nonpolar binding effects (Hubbard & Kirsch, 1972). With the present series of substituted N-benzoylimidazoles,p-CH3 is the only nonpolar substituent employed, and it will be noted in Figure 3 that it gives only a small deviation on the plots. The rate constants for the unsubstituted derivative I11 deviate slightly in a negative manner, which suggests that any steric interactions of the substituent groups with the active site are not appreciably increasing K, in comparison with that of 111. Even m-N02 gives a good fit on the plots established with the para substituents. Thus, it is probable that variations in K,, are not significantly altering the slopes. The p value at pH 7.5 very likely indicates that increased electron withdrawal in the acyl group of the N-benzoylimidazoles increases the rate constant kz for acylation. However, this p is considerably less positive than that in hydroxide ion catalyzed hydrolysis ( p = 1.4) (Fee & Fife, 1966a). The small value of p indicates only moderate charge development in the transition state (IX); it is probable n
I
H-N
I
9
N - - - H ---bCH,
I
IX
that neither bond making nor bond breaking has progressed to a great extent. The data are also in accord with a mechanism in which there is no C-N bond breaking in the transition state. In acylation reactions of substituted phenyl acetates (Bender & Nakamura, 1962) p is +1.8 for k2/K,,,, but the correlation was better with 0- than with u. Hubbard & Kirsch (1972) determined a p of 0.97 (k2/K,) for acylation of the enzyme by a series of acyl-substituted p-nitrophenyl benzoates and 1.6 for the analogous 2,4-dinitrophenyl benzoates. The p for OH--catalyzed hydrolysis of ethyl benzoates (Bruice & Benkovic, 1966) is large and positive (2.48), possibly reflecting the influence of electron withdrawal on nucleophilic attack by OH- at the deactivated carbonyl of the benzoate esters. The p for OH--catalyzed hydrolysis of p-nitrophenyl benzoates (Hubbard & Kirsch, 1972) is +2.0. The p for alkaline hydrolysis of substituted N-benzoylimidazoles is only 1.4, even though the imidazole anion leaving group has a pK,
V O L . 2 3 , N O . 1 3 , 1984
2987
of 14.5 (Walba & Isensee, 1961), Le., only slightly less than that of ethanol (Ballinger & Long, 1960). Likewise, the p value for water catalysis is 1.35 (Choi & Thornton, 1974). The small p values for hydrolysis of N-benzoylimidazoles must reflect transition-state structures that are quite different than those in comparable reactions of esters. General acid catalysis by the His-57 conjugate acid does not occur in the nucleophilic attack of Ser-195 on neutral N-acylimidazoles; a reaction involving a zwitterionic active site, i.e., His-57 conjugate acid and Ser-195 anion, can be ruled out with N-acylimidazoles having aliphatic acyl groups (Kogan et al., 1982) and also the N-benzoylimidazoles of the present study, because such a mechanism would require true secondorder rate constants greater than those for a diffusion-controlled reaction (1O'O M-* s-').~ A mechanism involving attack of the serine anion on the N-acylimidazole conjugate acid can be ruled out for the same reason and also because attack of the serine anion does not occur in the reactions of the N'methylated species at the pH values employed. General acid catalysis by the His-57 conjugate acid could occur in breakdown of a tetrahedral intermediate, but such a stepwise reaction would require that the tetrahedral intermediate be sufficiently stable to reach equilibrium with respect to proton transfer. A mechanism of that type is not in accord with a transition-state resembling reactants. It is probable that in the neutral species reactions His-57 is acting as a general base by partially abstracting a proton from Ser-195 in the transition state (IX). A general-base mechanism is required in acylation by the "-methylated compound (VIII). The position of a proton in the transition state of reactions of N-acylimidazoles can be determined by comparison of rate constants for reactions of exactly analogous "-methylated derivatives and unsubstituted compounds (Wolfenden & Jencks, 1961; Oakenfull et al., 1971). As with N-(3,3-dimethylbutyry1)-N'-methylimidazoliumion (Kogan et al., 1982) and in contrast to the nearly pH-independent acylation reactions of I-IV, the log (k2K,)-pH profile for acylation by N- [p-(dimethylamino) benzoyl]-N'-methylimidazolium ion (VIII) shows a rising arm with increasing pH with a slope of 1.O and a maximum in rate at pH >7. The pKappof 6.5 is near that expected for the conjugate acid of His-57. Thus, the k2/K,,, vs. pH profile indicates participation by the base form of histidine- 5 7. Histidine-57 is most likely participating in the acylation reaction as a proton-transfer agent. The D 2 0 solvent isotope effect on k2/K, in the reaction of VI11 is 2.1 while that in acylation of the enzyme by N-(3,3-dimethy1butyryl)-Ntmethylimidazolium ion is 3.1 (Kogan et al., 1982). Interpretation of D 2 0 solvent isotope effects in enzyme reactions is difficult (Jencks, 1963), but it has been argued that the solvent isotope effect in both acylation and deacylation reactions of a-chymotrypsin allows a reliable mechanistic interpretation (Bender et al., 1964; Bender & Hamilton, 1962). This viewpoint is strongly supported by the deacylation of [ (p-nitrophenoxy)carbonyl]chymotrypsin, which proceeds in large part via nucleophilic attack by His-57 with a DzOsolvent isotope effect close to unity as predicted for a nucleophilic reaction (Hutchins & Fife, 1972). It would reasonably be The ratio of zwitterionic to neutral active site would be approximately lo-' [see the equations in Kogan & Fife (1982)l. Therefore in view of values of k 2 / K mgreater than 10' M-' s-' for reaction of substituted N-benzoylimidazoles (11-VII), true second-order rate constants greater than 1Olo M-l s-I would be required for a reaction involving a zwitterionic active site in which Ser-195 is ionized and His-57 is present as the conjugate acid.
2988
BIOCHEMISTRY
expected that KmDl0 would be somewhat less than KmHz0 (Bender et al., 1964; Bender & Hamilton, 1962). A smaller k,/K,,, in D 2 0 than in H 2 0 therefore indicates that a proton is being transferred in the transition state. Since general-acid catalysis is precluded by the N'-methylated leaving group, this indicates that His-57 is participating by partially abstracting a proton from the Ser-195 hydroxyl group in the transition state (X).
The log (k2/Km)vs. pH profiles for the "-methylated and unmethylated analogues I and VI11 extrapolate to the same pH at approximately pH 4. Thus, since the "-methylated derivative is a good model, the reactions of the unmethylated N-benzoylimidazoles at low pH must involve the conjugate acids in which nitrogen is protonated. The values of k2/Km for the series of N-benzoylimidazoles are closely similar at pH 5, i.e., in contrast to the positive p at pH 7.5 that at pH 5 is nearly zero. At pH 5, the N-benzoylimidazoles will be significantly protonated (- 10%). Substituent effects on the pKa would be reasonably small (Choi & Thornton, 1974). It can be seen in Figure 4 that the pK,' for the dimethylaminosubstituted compound is 4; a pKa,, of 3.4 was determined in nonenzymatic hydrolysis reactions of the m-nitro derivative. Therefore, p for the protonation (KJ) is approximately 0.4. The p values near zero that have been found previously in hydronium ion catalyzed ester and amide hydrolysis (Bruice & Benkovic, 1966) are due to compensating effects of electron withdrawal on protonation and nucleophilic attack by water. However, the p value for hydronium ion catalyzed hydrolysis of substituted N-benzoylimidazoles at pH 4.7-6 is 0.99 (Choi & Thornton, 1974), which indicates that the ease of nucleophilic attack outweighs basicity considerations in the nonenzymatic hydrolysis of these compounds. The p near zero at pH 4-5 in the acylation reaction can be considered to reflect improvement of the leaving group by protonation, thereby reducing p by shifting the position of the transition state toward reactant. The p near zero is in accord with a transition state in which there is little difference in charge in comparison with the reactant. Thus, p implies a transition state with very little bond making with the nucleophile and little or no C-N bond breaking, with the transition state more closely resembling reactants than that for the reaction of the neutral species at pH 7.5. The values of k 2 / K mfor I and VI11 at pH 7 . 5 differ by a factor of only 150 even though the pK, of the leaving group of VI11 is over 7 pKa units more favorable (the pKa of N methylimidazole is 7). Likewise, there is only a difference of 155 in k2/Km values for N-(3,3-dimethylbutyryl)-N'methylimidazoliumion and N-( 3,3-dimethylbutyryl)imidazole (Kogan & Fife, 1982). The small rate ratio in these reactions is in accord with leaving groups that are similarly stabilized in the transition state and f or with reactions in which there is little bond breaking. It may be noted that the ratio of the rate constants koH' and koH is also similar in nonenzymatic OH--catalyzed hydrolysis of the unmethylated and N'methylated N-benzoylimidazoles I and VIII. The values of koH for OH--catalyzed hydrolysis differ by a factor of 470 with
KOGAN AND FIFE
the methylated and unmethylated N-acetyl derivatives (Wolfenden & Jencks, 1961; Jencks & Carriuolo, 1959). The similar rate constant ratios suggest that the same chemical factors are involved in the enzymatic acylation reactions and the nonenzymatic OH-catalyzed hydrolysis reactions of the N-benzoylimidazoles and the corresponding "-methylated compounds. However, the smaller p values in the enzymatic reactions show that there is very likely less bond formation in the transition state than in the hydrolysis reactions. An early transition state would result if the nucleophilic attack by Ser-195 is strongly facilitated by the enzyme. The ability of serine to act as a nucleophile must be due in part to the intracomplex nature of the attack step and to the highly efficient general-base catalysis by His-57 that is experimentally demonstrated in the acylation reactions of the "-methylated derivative VIII. Registry No. I, 87970-46-5; 11, 10364-93-9; 111, 10364-94-0; IV, 10347-1 1-2; V, 10364-95-1; VI, 87235-62-9; VII, 61652-82-2; VIII, 89 165-28-6; a-chymotrypsin, 9004-07-3.
References Ballinger, P., & Long, F. A. (1960) J . Am. Chem. SOC.82, 795. Bender, M. L., & Hamilton, G. A. (1962) J . Am. Chem. SOC. 84, 2570. Bender, M. L., & Nakamura, K. (1962) J . Am. Chem. SOC. 84, 2577. Bender, M. L., & Zerner, B. (1962) J . Am. Chem. SOC.84, 2550. Bender, M. L., & Kezdy, F. J. (1964) J . Am. Chem. SOC.86, 3704. Bender, M. L., Schonbaum, G. R., & Zerner, B. (1962) J. Am. Chem. SOC.84, 2562. Bender, M. L., Clement, G. E., Kezdy, F. J., & Heck, H. (1964) J . Am. Chem. SOC.86, 3680. Bernhard, S.A,, Lee,B. F., & Tashjian, Z. H. (1966) J. Mol. Biol. 18, 405. Brandt, K. G., Himoe, A., & Hess,G. P. (1967) J. Biol. Chem. 242, 3973. Brot, F. E., & Bender, M. L. (1969) J . Am. Chem. SOC.91, 7187.
Bruice, T. C., & Benkovic, S . (1966) Bioorganic Mechanisms, W. A. Benjamin, New York. Caplow, M., & Jencks, W. P. (1962) Biochemistry 1, 883. Choi, M., & Thornton, E. R. (1974) J . Am. Chem. SOC.96, 1428. Cunningham, L. W., & Brown, C. S . (1956) J . Biol. Chem. 221, 287. Doub, L., & Vandenbelt, J. M. (1947) J . Am. Chem. SOC.69, 2714. Fastrez, J., & Fersht, A. R. (1973) Biochemistry 12, 1067. Fee, J. A. (1967) Ph.D. Thesis, University of Southern California. Fee, J. A., & Fife, T. H. (1966a) J . Org. Chem. 31, 2343. Fee, J. A., & Fife, T. H. (1966b) J . Phys. Chem. 70, 3268. Fersht, A. (1977) Enzyme Structure and Mechanism, p 96, W. H. Freeman, San Francisco. Fife, T. H. (1965) J . Am. Chem. SOC.87, 4597. Fujita, T., Iwasa, J., & Hansch, C. (1964) J. Am. Chem. SOC. 86, 5175. Gutfreund, H., & Sturtevant, J. M. (1956) Biochem. J . 63, 656. Hammett, L. P. (1940) Physical Organic Chemistry, Chapter 7, McGraw-Hill, New York. Hammond, B. R., & Gutfreund, H. (1955) Biochem. J . 61, 187.
Biochemistry 1984, 23, 2989-2994 Himoe, A., Brandt, K. G., DeSa, R. J., & Hess, G. P. (1969) J . Biol. Chem. 244, 3483. Hubbard, C. D., & Kirsch, J. F. (1972) Biochemistry 11, 2483. Hutchins, J. E. C., & Fife, T. H. (1972) J . Am. Chem. SOC. 94, 8848. Inagami, T., York, S. S., & Patchornik, A. (1965) J . Am. Chem. SOC.87, 126. Jencks, W. P. (1963) Annu. Rev. Biochem. 32, 603. Jencks, W. P., & Carriuolo, J. (1959) J . Biol. Chem. 234, 1272. Kezdy, F. J., Clement, G. E., & Bender, M. L. (1964) J . Am. Chem. SOC.86, 3690. Kirsch, J. F., Clewell, W., & Simon, A. (1968) J. Org. Chem. 33, 127. Kogan, R. L., Fee, J. A., & Fife, T. H. (1982) J. Am. Chem. SOC.104, 3569. Laidler, K. J., & Barnard, M. L. (1956) Trans. Faraday SOC. 52, 497.
2989
Marini, J. L., & Caplow, M. (197 1) J . Am. Chem. SOC.93, 5560. Oakenfull, D. G., & Jencks, W. P. (1971) J . Am. Chem. SOC. 93, 178. Oakenfull, D. G.,Salvesen, K., & Jencks, W. P. (197 1) J. Am. Chem. SOC.93, 188. O'Leary, M. H., & Kluetz, M. D. (1972) J . Am. Chem. SOC. 94, 3585. Schonbaum, G. R., Zerner, B., & Bender, M. L. (1961) J . Biol. Chem. 236, 2930. Staab, H. A,, Luking, M., & Durr, F. H. (1962) Chem. Ber. 95, 1275. Walba, H., & Isensee, R. W. (1961) J . Org. Chem. 26,2789. Wolfenden, R., & Jencks, W. P. (1961) J . Am. Chem. SOC. 83, 4390. Zerner, B., & Bender, M. L. (1964) J . Am. Chem. SOC.86, 3669. Zerner, B., Bond, R. P. M., & Bender, M. L. (1964) J . Am. Chem. SOC.86, 3674.
Low-Temperature Reactions of Trypsin with p-Nitroanilide Substrates: Tetrahedral Intermediate Formation or Enzyme Isomerization+ Paul D. Compton and Anthony L. Fink*
ABSTRACT:
The reactions of trypsin with the p-nitroanilides N*-carbobenzoxy-L-arginine,and of N*-carbobenzoxy-L-lysine, N*-benzoyl-L-arginine have been studied in the 0 to -30 OC temperature region, over a range of pH* values, using 65% (v/v) aqueous dimethyl sulfoxide cryosolvent. At alkaline pH*, -30 OC, the catalytic reaction appears as a slow "burst" of product (or intermediate) followed by turnover. For all three substrates, the rate of the burst phase is identical. Preincubation of the enzyme at -30 OC abolishes the burst. On addition of trypsin to the cryosolvent at -30 OC, a timedependent decrease in fluorescence emission is observed with
the same rate as that of the burst with the anilides. The burst phase is thus interpreted as reflecting a temperature/solvent-induced isomerization of trypsin to a less catalytically efficient form, rather than the previously suggested formation of a tetrahedral intermediate [Compton, P. D., & Fink, A. L. (1980) Biochem. Biophys. Res. Commun. 93,427-4311. The isomerization is not observed at high temperature (20 "C)or at neutral pH*. The burst phase was not observed with aqueous methanol cryosolvent, indicating that it is sensitive to both cosolvent and temperature.
r e r e is considerable interest in the question of whether tetrahedral adducts exist as discrete intermediates or as transition states in protease catalysis. We have investigated the trypsin-catalyzed hydrolysis of p-nitroanilide substrates by using cryoenzymology in order to detect such putative intermediates. In an earlier study (Compton & Fink, 1980), we reported that a "burst" of absorbance in the 350-410-nm region could be observed prior to turnover in the reaction with Na-carbobenzoxy-L-lysinep-nitroanilide (ZLyspNA)' and attributed the reaction to the formation of a tetrahedral intermediate. Previous investigations (Fink, 1974) have shown that neither the catalytic nor the structural properties of trypsin are perturbed by exposure at 0 OC to 65% (v/v) dimethyl sulfoxide. However, the present structural (protein fluorescence emission) and catalytic (anilide hydrolysis) studies indicate that at lower temperatures in this cryosolvent a structural transition occurs resulting in reduced reactivity toward
anilide substrates. Trypsin has a long history of temperatureand ligand-induced transitions (Otero et al., 1980; Mares-Guia et al., 1981) to which the present isomerization must be added.
From the Division of Natural Sciences, University of California, Santa Cruz, California 95064. Received October 20, 1983. This research was supported by a grant from the National Science Foundation.
0006-2960/84/0423-2989$01.50/0
Experimental Procedures Materials. NWarbobenzoxy-L-lysine p-nitroanilide, lot F877, purified by preparative TLC, and the arginine analogue, lot 63797P, were obtained from Vega Biochemicals. The former was also obtained as a generous gift from Prof. Tesser, Organic Chemistry Department, Nijmegen, Holland. NaBenzoyl-L-arginine p-nitroanilide, lot 18C0224, was from Sigma; 3 times crystallized trypsin from Worthington, lot TRL 3 30D727, was used without further purification. Stock solutions of enzyme were prepared in l mM HC1 and stored at 4 "C. The activity was checked daily by burst titration with Abbreviations: TI, tetrahedral intermediate; ZLyspNA, NWarbobenzoxy-L-lysine pnitroanilide; ZArgpNA, N"-carbobenzoxy-L-arginine pnitroanilide; ZLyspNP, N"-carbobenzoxy-L-lysinep-nitrophenyl ester; BzArgpNA, N"-benzoyl-L-arginine p-nitroanilide; TLC, thin-layer chromatography.
0 1984 American Chemical Society
